npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@wastech-mdlint/mcp-server

v0.2.1

Published

MCP server for Markdown linting: read-only Model Context Protocol tools for lint findings, the context graph, and change impact.

Readme

@wastech-mdlint/mcp-server

A Model Context Protocol server that gives an agent read-only access to a repository's Markdown context: lint findings, the context graph, reachability slices, change impact, and a compiled skill.

stdio only, read-only, six tools. No HTTP or SSE transport, no mutating tools, no code-plugin execution. Every tool runs through @wastech-mdlint/core, the same pipeline the CLI uses — this package is the host adapter around it: tool registration, input validation, structured output, and error wrapping.

Run it

npx @wastech-mdlint/mcp-server        # run directly, no install
npm i -D @wastech-mdlint/mcp-server   # or install the wastech-mdlint-mcp bin

Add it to any stdio-based MCP host (Claude Code's .mcp.json, Claude Desktop's claude_desktop_config.json, and so on):

{
  "mcpServers": {
    "wastech-mdlint": {
      "command": "npx",
      "args": ["-y", "@wastech-mdlint/mcp-server"],
    },
  },
}

Node.js >=24.17.0. Readiness is announced on stderr; stdout carries only the protocol, so nothing the server logs can corrupt a JSON-RPC frame.

The six tools

| Tool | What it does | | --- | --- | | lint | Lint ad-hoc Markdown content against an explicit rule list. Does not load project config. | | lint-files | Lint the project's Markdown files using the resolved config, or the zero-config default. | | context-graph | Build the project's context graph — nodes, edges, cycles, or a summary document. | | context-slice | The files reachable within depth hops of a query resolved by exact match. | | impact-analysis | The blast radius of changing one file: direct and transitive dependents. | | compile-context | Compile the project skill (SKILL.md) from config.compile. |

All six carry a readOnlyHint annotation. Five return structuredContent alongside an outputSchema; compile-context returns two plain-text content blocks instead.

The five file-based tools accept a cwd. A cwd that does not exist, or that names a file rather than a directory, is rejected with an actionable error rather than answered with an empty result — an empty corpus is indistinguishable from a clean repository, so a plausible-looking success would answer a different question than the one asked. Omit it to analyze the server process's own working directory.

Errors

Tool failures use a structured { code, message, hint } contract, carried in structuredContent with message and hint also rendered into the text block, so a host that shows only text still sees the actionable half.

One limit is deliberate: that contract covers failures the tool itself detects. An argument shape the tool's advertised inputSchema rejects outright is refused by the protocol layer before the tool runs, and comes back as a validation error with no structuredContent at all.

Documentation

License

MIT — see LICENSE.